Angband.oook.cz
Angband.oook.cz
AboutVariantsLadderForumCompetitionComicScreenshotsFunniesLinks

Go Back   Angband Forums > Angband > Vanilla

Reply
 
Thread Tools Display Modes
Old April 11, 2011, 17:52   #11
Tiburon Silverflame
Swordsman
 
Join Date: Feb 2010
Posts: 405
Tiburon Silverflame is on a distinguished road
IIRC, the old RNG was a linear congruential...and it compares *horribly* to MT, in a proper statistical analysis. Sure, over 10,000 trials, they probably give similar results...but LC is known to give particular shorter-term patterns.

You certainly can't blame every double failure on the RNG...and with combat spellcasting, it's probably true that you can blame only a few, unless the "spell success" method has an isolated RNG all to itself. If it doesn't, then in combat situations you'll have attack and damage rolls, presumably using the same RNG, and therefore your next spellcasting attempt won't be "consecutive" from the standpoint of the generated numbers.
Tiburon Silverflame is offline   Reply With Quote
Old April 12, 2011, 07:17   #12
zaimoni
Knight
 
zaimoni's Avatar
 
Join Date: Apr 2007
Posts: 590
zaimoni is on a distinguished road
Quote:
Originally Posted by Tiburon Silverflame View Post
IIRC, the old RNG was a linear congruential...and it compares *horribly* to MT, in a proper statistical analysis.
The old complex RNG was/is an addition-shift initialized by a linear-congruential.

Just about anything other than Mother-of-All compares horribly to Mersenne Twister for non-cryptographic purposes.
Quote:
Originally Posted by Tiburon Silverflame View Post
If it doesn't, then in combat situations you'll have attack and damage rolls, presumably using the same RNG, and therefore your next spellcasting attempt won't be "consecutive" from the standpoint of the generated numbers.
Which is why the important measure is correlation across phase shifts (which is precisely where linear-congruential is awful -- the best known LCG as of 2004 has detectable strong correlation at sampling interval 8, but not intervals 2..7).

By contrast, the old RNG was not empirically testable [on my end] as problematic for events with probability greater than 1/27.
__________________
Zaiband: end the "I shouldn't have survived that" experience. V3.0.6 fork on Hg.
Zaiband 3.0.10 ETA Mar. 7 2011 (Yes, schedule slipped. Latest testing indicates not enough assert() calls to allow release.)
Z.C++: pre-alpha C/C++ compiler system (usable preprocessor). Also on Hg. Z.C++ 0.0.10 ETA December 31 2011
zaimoni is offline   Reply With Quote
Old April 12, 2011, 07:32   #13
zaimoni
Knight
 
zaimoni's Avatar
 
Join Date: Apr 2007
Posts: 590
zaimoni is on a distinguished road
Quote:
Originally Posted by Timo Pietilš View Post
If I understood correctly that would be 18 consecutive successes, not failures. Failures would be 0.33^18 ~= 2.154*10^-9 which is about 1 in 464246974
Yes, agreed.
__________________
Zaiband: end the "I shouldn't have survived that" experience. V3.0.6 fork on Hg.
Zaiband 3.0.10 ETA Mar. 7 2011 (Yes, schedule slipped. Latest testing indicates not enough assert() calls to allow release.)
Z.C++: pre-alpha C/C++ compiler system (usable preprocessor). Also on Hg. Z.C++ 0.0.10 ETA December 31 2011
zaimoni is offline   Reply With Quote
Old April 12, 2011, 08:09   #14
Mondkalb
Knight
 
Mondkalb's Avatar
 
Join Date: Apr 2007
Posts: 936
Mondkalb is on a distinguished road
I have learned (through many many deaths of characters) that a failure rate of 5% means: Don't rely on it in critical situations. It will fail four or five times in a row, if necessary to kill your character. ^^
Mondkalb is offline   Reply With Quote
Old April 12, 2011, 20:42   #15
Max Stats
Swordsman
 
Join Date: Jun 2010
Posts: 324
Max Stats is on a distinguished road
I'm glad I'm not the only one who thinks repeated failures are too common. I always wondered if the RNG was susceptible to occasional lack of independence between successive numbers. I assumed I was just noticing the failures and ignoring the successes, but maybe there is something to it.
Max Stats is offline   Reply With Quote
Old April 13, 2011, 05:14   #16
d_m
Angband Devteam member
 
d_m's Avatar
 
Join Date: Aug 2008
Location: Philadelphia, PA, USA
Age: 38
Posts: 1,516
d_m is on a distinguished road
I'm no expert in this, but I got permission to include the WELL RNG in Angband under the GPL which is what we're using now. See this page:

http://www.iro.umontreal.ca/~panneton/WELLRNG.html

In particular, if you look at the WELL1024a.c implementation, we're using a version of that (slightly modified to return integral values instead of floating-point values).
__________________
linux->xterm->screen->pmacs
d_m is offline   Reply With Quote
Old April 13, 2011, 14:28   #17
Camcolit
Scout
 
Join Date: Apr 2011
Posts: 30
Camcolit is on a distinguished road
Just noticed all these replies now.

Quote:
Originally Posted by EpicMan View Post
Except that the older rng, which I believe is the one Zangband has, has some issues. One failure seems to make successive failures likely. That's why they changed the RNG in Vanilla.So your initial failure rate may be 5%, but a failure will make a second one much more likely.
This seems to fit what's happening in my game. I don't mind all that much. In fact from a RP point of view it makes sense, if you fail a spell maybe your confidence is shot and are less likely to succeed the next try, especially when some abomination is trying to chop you up into neat cubes .
Camcolit is offline   Reply With Quote
Old April 14, 2011, 05:53   #18
Philip
Knight
 
Join Date: Jul 2009
Location: Prague,Czech Republic
Posts: 894
Philip is on a distinguished road
Quote:
Originally Posted by Camcolit View Post
Just noticed all these replies now.



This seems to fit what's happening in my game. I don't mind all that much. In fact from a RP point of view it makes sense, if you fail a spell maybe your confidence is shot and are less likely to succeed the next try, especially when some abomination is trying to chop you up into neat cubes .
Yep, Angband is the only game with OCD nasties.
Philip is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
320 stair failure dos350 AAR 43 February 1, 2011 01:18
reduce artifact drop rate fizzix Vanilla 90 October 13, 2010 15:41
Furyband Gold Spell Storing Question Theyli Variants 3 April 17, 2010 20:41
Rings of escaping affect spell failure?? Magnate Vanilla 13 May 25, 2009 03:55
Compile failure for 3.0.9 on Mac OS X 10.4.10 TreeFrog Vanilla 4 November 8, 2007 14:58


All times are GMT +1. The time now is 04:32.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2019, vBulletin Solutions Inc.